U.S. patent application number 13/025431 was filed with the patent office on 2011-09-01 for deposition rate control.
This patent application is currently assigned to First Solar, Inc.. Invention is credited to Markus E. Beck, Ashish Bodke, Ulrich A. Bonne, Raffi Garabedian, Erel Milshtein, Ming L. Yu.
Application Number | 20110212256 13/025431 |
Document ID | / |
Family ID | 44368133 |
Filed Date | 2011-09-01 |
United States Patent
Application |
20110212256 |
Kind Code |
A1 |
Beck; Markus E. ; et
al. |
September 1, 2011 |
DEPOSITION RATE CONTROL
Abstract
An vapor deposition control system includes a multi-level
control scheme.
Inventors: |
Beck; Markus E.; (Scotts
Valley, CA) ; Yu; Ming L.; (Fremont, CA) ;
Milshtein; Erel; (Cupertino, CA) ; Bodke; Ashish;
(San Jose, CA) ; Bonne; Ulrich A.; (Sunnyvale,
CA) ; Garabedian; Raffi; (Los Altos, CA) |
Assignee: |
First Solar, Inc.
Perrysburg
OH
|
Family ID: |
44368133 |
Appl. No.: |
13/025431 |
Filed: |
February 11, 2011 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61304058 |
Feb 12, 2010 |
|
|
|
Current U.S.
Class: |
427/10 ; 118/665;
118/667; 118/688; 427/8; 427/9 |
Current CPC
Class: |
C23C 14/0623 20130101;
C23C 14/548 20130101; G01N 21/3103 20130101; C23C 14/543 20130101;
G01N 21/211 20130101; G01N 21/8422 20130101; G01N 21/359 20130101;
C23C 14/544 20130101; G01N 2021/8416 20130101; C23C 14/545
20130101; G01N 21/3563 20130101 |
Class at
Publication: |
427/10 ; 427/8;
427/9; 118/688; 118/667; 118/665 |
International
Class: |
C23C 16/52 20060101
C23C016/52; C23C 16/455 20060101 C23C016/455 |
Claims
1. A method of controlling a vapor deposition rate and composition
comprising: measuring a vapor flux rate of a vapor being fed from a
vapor source and deposited; calculating a deposition rate based on
the measured vapor flux rate, wherein a correlation function
between flux rate and the deposition rate is used to calculate the
deposition rate; and controlling the deposition rate by a feedback
control loop based on the deposition rate.
2. The method of claim 1, further comprising: measuring a vapor
source temperature of the vapor source; and controlling the
deposition rate by a first check control loop, wherein the first
check control loop comprises a correlation function between
deposition rate and vapor source temperature to verify the
calculated deposition rate by using the measured vapor source
temperature.
3. The method of claim 2, further comprising: measuring a vapor
source power of the vapor source; and controlling the deposition
rate by a second check control loop, wherein the second check
control loop comprises a correlation function between deposition
rate and vapor source power to verify the calculated deposition
rate by using the measured vapor source power.
4. The method of claim 1, further comprising: measuring a vapor
source temperature of the vapor source; and controlling the
deposition rate by a first check control loop, wherein the first
check control loop comprises a correlation function between flux
rate and vapor source temperature to verify the measured vapor flux
rate by using the measured vapor source temperature.
5. The method of claim 1, further comprising establishing a target
deposition layer thickness of the deposited vapor.
6. The method of claim 4, further comprising setting the vapor flux
rate based on the target deposition layer thickness.
7. The method of claim 4, further comprising measuring the
deposited film thickness during deposition.
8. The method of claim 6, further comprising comparing the measured
deposited film thickness to the target deposition layer thickness
and controlling the deposition rate by a feedback control loop
based on the measured deposition film thickness.
9. The method of claim 6, wherein the deposition film thickness is
measured using one or more of near infrared reflectometer, X-ray
fluorescence sensor, ellipsometer, light scattering sensor, or
optical transmission sensor.
10. The method of claim 6, wherein the deposition film thickness is
measured using an in-situ instrument to monitor the deposition
process in real time.
11. The method of claim 4, further comprising the steps of
adjusting the vapor flux rate and iterating until the target
deposition layer thickness is present.
12. The method of claim 6, wherein the step of measuring the
deposition film thickness comprises timing the deposition film
thickness measurement to occur after the step of measuring the flux
rate.
13. The method of claim 6, wherein the step of measuring the
deposition film thickness comprises timing the deposition film
thickness measurement to occur after vapor has been deposited.
14. The method of claim 1, wherein the step of measuring the vapor
flux comprises using an atomic absorption spectrometer, an electron
impact emission spectrometer, an ion gauge, optionally in a
configuration enabling the monitor to measure the position
sensitive flux rate.
15. A vapor deposition rate control system comprising: a vapor flux
monitor capable of measuring a vapor flux rate of a vapor being
deposited; a vapor flux control module capable of reading the flux
monitor and controlling the vapor flux rate by adjusting a vapor
source feed rate from a vapor source; and a feedback control loop
based on a correlation function between the flux rate and a
deposition rate of the vapor being deposited, to correlate the flux
rate to the deposition rate and control the deposition rate by the
control module.
16. The vapor deposition rate control system of claim 15, further
comprising: a vapor source temperature sensor capable of measuring
a vapor source temperature of the vapor source; and a first check
control loop comprising a correlation function between deposition
rate and vapor source temperature to compare the deposition rate
correlated to the vapor source temperature with the deposition rate
correlated to the measured flux rate.
17. The vapor deposition rate control system of claim 16, wherein
the control system further comprises: a vapor source power sensor
capable of measuring a vapor source power of the vapor source; and
a second check control loop comprising a correlation function
between flux rate and vapor source power to compare the deposition
rate correlated to the vapor source power with the deposition rate
correlated to the measured flux rate.
18. The vapor deposition rate control system of claim 15, further
comprising: a vapor source temperature sensor capable of measuring
a vapor source temperature of the vapor source; and a first check
control loop comprising a correlation function between flux rate
and vapor source temperature to compare the flux rate correlated to
the vapor source temperature with the measured flux rate.
19. The vapor deposition rate control system of claim 15, further
comprising a data storage apparatus storing a target deposition
layer thickness of a deposited vapor, wherein the data storage
apparatus comprises a self-teaching algorithm to allow selection of
the vapor flux rate as a function of the target deposition layer
thickness.
20. The vapor deposition rate control system of claim 15, further
comprising a film thickness monitor capable of measuring the
thickness of a deposited vapor, wherein the film thickness monitor
comprises an in-situ configuration when measuring the thickness of
a deposited vapor.
21. The vapor deposition rate control system of claim 20, wherein
the film thickness monitor comprises one or more of near infrared
reflectometer, X-ray fluorescence sensor, ellipsometer, light
scattering sensor, or optical transmission sensor.
22. The vapor deposition rate control system of claim 20, wherein
the film thickness monitor can monitor the deposition process in
real time.
23. The vapor deposition rate control system of claim 20, further
comprising a film thickness control module capable of adjusting the
vapor flux rate and iterating until the target deposition layer
thickness is present.
24. The vapor deposition rate control system of claim 23, wherein
the film thickness monitor measures deposition layer thickness
after the flux rate is measured.
25. The vapor deposition rate control system of claim 22, wherein
the vapor flux monitor comprises one or more of atomic absorption
spectrometer, electron impact emission spectrometer, or ion gauge
in a configuration enabling the monitor to measure the position
sensitive flux rate.
Description
CLAIM OF PRIORITY
[0001] This application claims priority to U.S. Provisional Patent
Application No. 61/304,058, filed on Feb. 12, 2010, which is
incorporated by reference in its entirety.
TECHNICAL FIELD
[0002] This invention relates to a material deposition process,
which can include a deposition rate control system.
BACKGROUND
[0003] Evaporation is a common method of thin film deposition. The
source material can be evaporated under reduced pressure, such as
in a vacuum. The vacuum allows vapor flux to travel directly to the
target object, where it condenses back to a solid state.
Evaporation is used in microfabrication, and to make macro-scale
products such as solar cell or metalized plastic film. Controlling
the deposition rates from evaporation sources can prove to be
difficult, in particular in an ambient environment of background
elements or for wide rate ranges.
DESCRIPTION OF DRAWINGS
[0004] FIG. 1 is a schematic showing the one-stage multilevel
thermal evaporation deposition control system.
[0005] FIG. 2 illustrates a setup of optical elements of a
deposition control system.
[0006] FIG. 3 illustrates a receiving end setup of optical elements
of the flux rate monitor.
[0007] FIG. 4 is a diagram of an exemplary photodiode output
waveform.
[0008] FIG. 5 is a schematic showing the thermal evaporation
deposition process of the CIGS layer.
[0009] FIG. 6 illustrates a three-stage multilevel control
scheme.
[0010] FIG. 7 illustrates a three-stage multilevel control
scheme.
[0011] FIG. 8 illustrates a configuration of a near infrared
reflectometry sensor with an in-situ configuration for in-line
deposition process.
[0012] FIG. 9 illustrates a configuration of an X-ray fluorescence
sensor with an in-situ configuration for in-line deposition
process.
DETAILED DESCRIPTION
[0013] In evaporation deposition, a metal source can be heated by
certain methods, such as passing a current through a container or
by focusing an electron beam on the metal's surface. As metal
evaporates, it forms a vapor flux that condenses on the cooler
surface of the target object (e.g. substrates) to form a thin film.
Evaporation is widely used in microfabrication, and to make
macro-scale products such as solar cell or metalized plastic film.
Controlling the deposition rates from evaporation sources can prove
to be difficult, in particular in an ambient environment of
background elements or compounds or for wide rate ranges. An
evaporation rate control system with multi-level control approach
is developed for thin film deposition process.
[0014] If various elements or compounds are coevaporated, species
sensitive information and compositional control are both desirable.
Various rate monitor technologies exist: some are species
sensitive, while others are not. Not all of these sensors are
equally capable to operate under a wide range of ambient conditions
(atmosphere, temperature). Furthermore, in-situ measurement result
of the flux rate is most likely not equal to the deposition rate on
the substrate. This can result from the material being deposited
having a sticking coefficient dictating that less than all the
atoms of the material that impinge on a substrate will be
deposited. While for well known conditions it may be possible to
establish correlation factors of flux rate versus deposition rate,
fluctuations in the process conditions (intentional or
unintentional) cannot be detected, nor compensated for. Thus,
design principles would necessitate solutions of inherently stable
sources, but this is not always possible or desirable. At the same
time, a control scheme needs to be capable of detecting and
reacting to fluctuations in the process condition to compensate for
variations in the flux to deposition rate relationship. The
evaporation rate control system disclosed herein addresses this
problem via a control loop approach.
[0015] In one aspect, a method of controlling a vapor deposition
rate and composition includes measuring a vapor flux rate of a
vapor being fed from a vapor source and deposited and calculating a
deposition rate based on the measured vapor flux rate. A
correlation function between flux rate and the deposition rate can
be used to calculate the deposition rate. The method can include
controlling the deposition rate by a feedback control loop based on
the deposition rate.
[0016] The method can include the steps of measuring a vapor source
temperature of the vapor source and controlling the deposition rate
by a first check control loop. The first check control loop can
include a correlation function between vapor source temperature and
deposition rate which can be used to verify the calculated
deposition rate by using the measured vapor source temperature. The
method can include the steps of measuring a vapor source power of
the vapor source and controlling the deposition rate by a second
check control loop. The second check control loop can include a
correlation function between vapor source power and deposition rate
to verify the calculated deposition rate by using the measured
vapor source power. The method can include the steps of measuring a
vapor source temperature of the vapor source and controlling the
vapor flux rate by a first check control loop. The first check
control loop can include a correlation function between flux rate
and vapor source temperature which can be used to verify the
measured vapor flux rate by using the measured vapor source
temperature.
[0017] The method can include establishing a target deposition
layer thickness of the deposited vapor. The method can include
setting the vapor flux rate based on the target deposition layer
thickness. The method can include measuring the deposited film
thickness during deposition. The method can include comparing the
measured deposited film thickness to the target deposition layer
thickness and controlling the deposition rate by a feedback control
loop based on the measured deposition film thickness. The
deposition film thickness can be measured using a near infrared
reflectometer. The deposition film thickness can be measured using
an X-ray fluorescence sensor. The deposition film thickness can be
measured using an ellipsometer. The deposition film thickness can
be measured using a light scattering sensor. The deposition film
thickness can be measured using an optical transmission sensor. The
deposition film thickness can be measured using an in-situ
instrument to monitor the deposition process in real time.
[0018] The method can include the steps of adjusting the vapor flux
rate and iterating until the target deposition layer thickness is
present. Measuring the deposition film thickness can include timing
the deposition film thickness measurement to occur after the step
of measuring the flux rate. Measuring the deposition film thickness
can include timing the deposition film thickness measurement to
occur after vapor has been deposited. Measuring the vapor flux can
include using an atomic absorption spectrometer. Measuring the
vapor flux can include using an electron impact emission
spectrometer. Measuring the vapor flux can include using an ion
gauge. Measuring the vapor flux can include using a configuration
enabling the monitor to measure the position sensitive flux
rate.
[0019] A vapor deposition rate control system can include a vapor
flux monitor capable of measuring a vapor flux rate of a vapor
being deposited, a vapor flux control module capable of reading the
flux monitor and controlling the vapor flux rate by adjusting a
vapor source feed rate from a vapor source, and a feedback control
loop. The feedback control loop can be based on a correlation
function between the flux rate and a deposition rate of the vapor
being deposited, to correlate the flux rate to the deposition rate
and control the deposition rate by the control module.
[0020] The vapor deposition control system can include a vapor
source temperature sensor capable of measuring a vapor source
temperature of the vapor source, and a first check control loop.
The first check control loop can include a correlation function
between vapor source temperature and deposition rate to compare the
deposition rate correlated to the vapor source temperature with the
deposition rate correlated to the measured flux rate. The system
can include a vapor source power sensor capable of measuring a
vapor source power of the vapor source, and a second check control
loop. The second check control loop can include a correlation
function between vapor source power and deposition rate to compare
the deposition rate correlated to the vapor source power with the
deposition rate correlated to the measured flux rate. The vapor
deposition control system can include a vapor source temperature
sensor capable of measuring a vapor source temperature of the vapor
source, and a first check control loop. The first check control
loop can include a correlation function between vapor source
temperature and flux rate to compare the flux rate correlated to
the vapor source temperature with the measured flux rate.
[0021] The vapor deposition rate control system can include a data
storage apparatus storing a target deposition layer thickness of a
deposited vapor. The data storage apparatus can include a
self-teaching algorithm to allow selection of the vapor flux rate
as a function of the target deposition layer thickness. The system
can include film thickness monitor capable of measuring the
thickness of a deposited vapor. The film thickness monitor can
include an in-situ configuration when measuring the thickness of a
deposited vapor. The film thickness monitor can include a near
infrared reflectometer. The film thickness monitor can include an
X-ray fluorescence sensor. The film thickness monitor can include
an ellipsometer. The film thickness monitor can include a light
scattering sensor. The film thickness monitor can include an
optical transmission sensor. The film thickness monitor can monitor
the deposition process in real time.
[0022] The vapor deposition rate control system can include a film
thickness control module capable of adjusting the vapor flux rate
and iterating until the target deposition layer thickness is
present. The film thickness monitor can measure deposition layer
thickness after the flux rate is measured. The vapor flux monitor
can include an atomic absorption spectrometer. The vapor flux
monitor can include an electron impact emission spectrometer. The
vapor flux monitor can include an ion gauge. The vapor flux monitor
can be configured to enable the monitor to measure the position
sensitive flux rate.
[0023] Referring to FIG. 1, in certain embodiments, a one-stage
multilevel thermal vapor deposition control system can include a
control module. The control system can include a vapor flux
monitor, a vapor source temperature sensor, and a vapor source
power sensor as first level sensors. The vapor flux monitor can
include optical elements 70, 80 used to measure the vapor flux rate
of the vapor being deposited. The control system can use
correlation functions of flux rate versus deposition rate to engage
a feedback control loop via flux monitor's measurement 10. Because
the material being deposited can have a sticking coefficient of
less than 1.00 (in which case, for example, less than all atoms
impinging on the deposition surface are deposited), a measured flux
rate can be multiplied by the sticking coefficient (among other
suitable calculations) to determine a deposition rate.
[0024] The evaporation rate control system can include one or more
check control loops to evaluate and/or refine the deposition rate
determined based on the measured flux rate, and the methodology for
calculating the deposition rate. For example, the evaporation rate
control system can use the vapor source temperature sensor as part
of a first check control loop, wherein a correlation function
between the vapor source temperature 20 and the deposition rate can
be used to verify the deposition rate calculated based on the
measured flux rate. The evaporation rate control system can include
additional or alternate check control loops. For example, the
evaporation rate control system can use the vapor source power
sensor as part of a second check control loop, wherein a
correlation function between deposition rate and vapor source power
30 can be used to verify the deposition rate calculated based on
the measured vapor flux rate. The evaporation rate control system
can include a check control loop which correlates flux rate and
vapor source temperature to verify the measured vapor flux rate by
using the vapor source temperature 20 measured by the vapor source
temperature sensor.
[0025] In some embodiments, a second level film thickness sensor
can be applied in-situ during the vapor deposition to send
deposited film thickness measurement 40 to the control module. When
the film thickness supervisory sensor detects a discrepancy to a
desired target deposition layer thickness, it can adjust the vapor
flux rate and iterate until the target deposition layer thickness
is achieved. The vapor deposition control system can use any
suitable tool, instrument, or method or combination or tools,
instruments, or methods to measure the film thickness of the
deposited vapor layer. For example, an X-ray fluorescence sensor
(XRF) can be used to measure the film thickness of the deposited
layer. X-ray fluorescence sensor (XRF) can include an energy
dispersive spectrometer (EDS). The energy dispersive spectrometer
can detect the emission of characteristic "secondary" (or
fluorescent) X-rays from a material that has been excited by
bombarding with high-energy X-rays (or gamma rays). By analyzing
the emission, the energy dispersive spectrometer can also provide
compositional information of the deposited layer on the substrates.
A near infrared reflectometer (NIR) can be used to measure film
thickness and can be included in the film thickness monitor.
[0026] The vapor deposition control system can use the control
module to adjust vapor source power 50 or continuous feed thermal
evaporation sources 60. In some embodiments, the integrated signals
from the sensors allow direct access to the material
consumption/mass loss of the source. Therefore, it can be used to
control the feedstock replenishment mechanism controlling the
source to a fixed fill level. It can also be used to control the
vapor source power to a fixed evaporation level. In some
embodiments, the vapor deposition control system can control the
deposition rate of an in-line, multi-stage deposition system
capable of continuous processing of moving substrates. Several
evaporation sources can be used in an in-line configuration in the
system.
[0027] The vapor flux rate monitor can include an atomic absorption
spectrometer or any other suitable sensors (e.g. electron impact
emission spectrometer or ion gauges). When an excited atom
de-excites, it emits a photon of characteristic wavelength. Atomic
absorption (AA) is the reverse of this process. If a beam of light
at the characteristic wavelength passes through a cloud of atoms
with uniform density across the light beam diameter, photons will
be absorbed by the atoms. The amount of absorption will depend on
the number of atoms in the light path.
[0028] For example, let
[0029] I.sub.in=incident light intensity
[0030] I.sub.out=transmitted light intensity
[0031] N=number of atoms the beam interacts.
[0032] Beer's law states that
I.sub.out=I.sub.inexp(-N/.alpha.)
[0033] .alpha. is a constant that is related to the cross-section
of optical absorption.
N=.alpha. ln(I.sub.in/I.sub.out).
[0034] The ratio I.sub.out/I.sub.in is directly related N.
Therefore AA can be used as a monitor of metal flux. Note that
dI.sub.out/dN=-(I.sub.out/.alpha.).
[0035] AA is most sensitive when I.sub.out is large, or when N is
small. The major characteristic absorption lines of Cu, In, and Ga
are shown below:
[0036] Copper 324.75(s)/327.40/217.90 nm
[0037] Gallium 287.4/294.36(s)/403.3 nm
[0038] Indium 303.94(s)/325.61 nm
`s` means the strongest. Therefore, the working wavelength of the
light source can be in the 280-330 nm range in the UV.
[0039] In some embodiments, the light source for AA can include the
Hollow Cathode Lamp (HCL) where an Ar or Ne plasma excites the
atoms of interest to emit the characteristic lines. The light
source for AA can include a tunable laser. There are diode lasers
in the wavelengths of interest. The emission from HCL can include a
background of many other emission lines. A bandpass filter either
at the source or at the detector can be used to filter the unwanted
emission lines. The intensity ratio I.sub.out/I.sub.in is needed
for the measurement. I.sub.in can be measured by shuttering off the
metal vapor. Another detector can be used to monitor the HCL
output. Therefore, corrections can be made to the light intensity
if necessary. The transmission of the optical system may change due
to metal deposition on the optical elements. If the metal vapor can
be shuttered off to measure I.sub.in, it will be equivalent to an
intensity change of the light source and be used as above. A
"white" light source, i.e., one that is not absorbed or scattered
by the metal vapor can also be used to monitor the change in the
transmission.
[0040] Referring to FIG. 2, as a sending end setup of optical
elements of the system (70 in FIG. 1 and FIG. 2), a hollow cathode
lamp can remain on during the experiment to maintain stability. The
broad (about 1''-1.5'') emission from the HCL is coupled through a
collimating lens to the optical fiber. The light is split, part to
the light intensity monitoring detector PD2 (if necessary), and
part for flux measurement (A). A fiber optic attenuator can be used
to adjust the light intensity. The white LED can be turned on and
off by the computer. It is also split between the light intensity
monitoring detector PD2 and for measurement (A). A mechanical
shutter can be included to eliminate the necessity of a white light
path.
[0041] With shuttering capability, both I.sub.in and I.sub.out can
be obtained with a single detector. As mentioned above, the HCL has
a background besides the main emission line. Referring to FIG. 3
showing a receiving end setup of optical elements of the flux rate
monitor (80 in FIG. 1 and FIG. 3), a UV bandpass filter can be
positioned in front of the detector. The detector can be a Si
photodiode. The light coming out from vapor flux will be
transported through the evaporation chamber wall by optical fiber
vacuum feedthrough. The output will be collimated by a small
collimating lens onto a silicon photodiode PD1 outside the
chamber.
[0042] In one embodiment, a procedure for metal flux measurement
includes the following steps: [0043] 1. Set HCL optical attenuator
to zero transmission. Read the background signal I.sub.back.
I.sub.back is from stray light and the dark current of the
detector. [0044] 2. With shutter on (no metal flux), adjust the
fiber optic attenuator so that the silicon photodiode output is
within the linear range of the detector. The signal
I.sub.shut=I.sub.in+I.sub.back. [0045] 3. Open the shutter (metal
flux on). The signal is I.sub.open=I.sub.out+I.sub.back.
[0046] From Beer's law:
N=.alpha. ln [(I.sub.shut-I.sub.back)/(I.sub.open-I.sub.back)].
[0047] N can not be obtained directly without knowing the
proportionality constant .alpha.. This is not necessary for the
purpose of flux control. The value .alpha. can be measured, for
example, by measuring the thickness of the deposited film.
Referring to FIG. 4, an exemplary photodiode output waveform is
shown. As shown in FIG. 3, the photodiode output waveform can be
idealistic square wave function in case of shuttered flux, for
example, in a system including a fast shutter to shutter the flux
resulting substantially in only either an on state or an off state.
In some embodiments, a slower shutter can be used, resulting in a
sloped output waveform.
[0048] In the case when the metal flux can not be shuttered, the
light intensity monitoring detector and the "white" LED can be
included in the flux rate monitor to calibrate out any change in
the transmission in the AA optical path, for example, due to metal
deposition on the optical elements. The light intensity monitoring
detector can be called PD2. The AA signal photodiode can be called
PD1. The calibration is needed between PD1 and PD2 for both the AA
sensing light and the white LED, assuming that the background
signals from the two PDs are always subtracted already.
[0049] In certain circumstances, procedure for calibration can be:
[0050] 1. Metal source off [0051] 2. Turn on HCL until stable
[0052] 3. With white LED off or LED fiber attenuator set to zero
output, set the HCL light level to a convenient value for
measurement using the HCL fiber attenuator [0053] 4. Read PD1 and
PD2 outputs [0054] 5. Initial calibration factor k.sub.HCLi=PD1
output/PD2 output [0055] 6. Take out the UV bandpass filters for
the photodiode detectors [0056] 7. With HCL fiber attenuator set to
zero output, set the LED light level to a convenient value for
measurement using the LED fiber attenuator [0057] 8. Read PD1 And
PD2 outputs [0058] 9. Initial calibration factor k.sub.LEDi=PD1
output/PD2 output
[0059] In some embodiments, during metal flux measurement, the HCL
can be kept on to maintain its stability. Therefore, PD1 can always
have an output I.sub.AA as the AA signal. However, the transmission
of the optics may change due to metal deposition on the optics
elements. Therefore, the correction needs to be made. For this
measurement, the white LED whose light may not be attenuated to any
significant extent by the metal flux can be used. The procedure to
make transmission correction can be:
[0060] 1. Block off the HCL light
[0061] 2. Take out the UV bandpass filters
[0062] 3. LED on
[0063] 4. Take reading: [0064] a. PD1: I.sub.LED1 [0065] b. PD2:
I.sub.LED2
[0066] 5. Take the ratio k.sub.LED=I.sub.LED1/I.sub.LED2
[0067] 6. Transmission correction factor
.beta.=k.sub.LEDi/k.sub.LED
[0068] 7. The transmission correction factor has to be applied to
the AA signal.
[0069] 8. Turn off LED
[0070] 9. Unblock HCL
[0071] 10. Replace UV bandpass filters
[0072] 11. Take reading: [0073] a. PD1: I.sub.AA [0074] b. PD2:
I.sub.HCL
[0074] I.sub.out=I.sub.AA
I.sub.in=I.sub.HCL*k.sub.HCLi/.beta..
[0075] Therefore
N = .alpha. ln ( I i n / I out ) = .alpha. ln ( I HCL * k HCLi /
.beta. I AA ) = .alpha. ln [ ( PD 2 signal / PD 1 signal ) * ( k
HCLi / .beta. ) ] ##EQU00001##
[0076] Here .alpha. has the same value as the .alpha. for the
shuttered case.
[0077] The above-described control loop can only guarantee that the
flux rate will be constant, but can be blind with respect to
external condition changes that impact the deposition rate, such as
sticking coefficient, background species, substrate temperature
variations, or source power fluctuation. When various elements or
compounds are coevaporated, such as CIGS film deposition, species
sensitive information and compositional control are also
desirable.
[0078] Photovoltaic devices using copper indium (di) selenide (CIS)
and their alloys with gallium (CIGS) can be manufactured using a
variety of techniques. Materials can be co-evaporated.
Co-evaporation of CIGS thin film via two-stage and three-stage
processes has been widely used. Referring to FIG. 5, CIGS film
thermal evaporation system 100 can include chamber 110. Chamber 110
can be connected to a vacuum system which allows working at
pressures of about 10.sup.-6 Torr. System 100 can include any
suitable number of boats (e.g., three or four boats used to
evaporate Se, In, (Ga), and Cu, respectively) and thickness monitor
160 with quartz crystal sensor 150, which was used for measuring
the flux rate of the evaporated elements. System 100 can include
programmable power source and related controller 140. Substrate 120
can be mounted on mounting fixture 130 or positioned in any other
suitable manner. System 100 can further include any suitable
substrate heating module if necessary. Mounting fixture 130 can be
rotary and hold substrate 120 facing down. Evaporation processes
can include a plurality of stages and species. In some embodiments,
the CIGS deposition system can be an in-line, 3 stage deposition
system capable of continuous processing of moving substrates.
Several evaporation sources can be used in an in-line configuration
in the system.
[0079] To monitor and control the rate and composition of
deposition process (e.g. CIGS film deposition), the vapor
deposition control system can use a multi-level control approach
for thin film deposition process. The vapor deposition control
system can include a desired target deposition layer thickness for
the respective element or compound deposited. The target layer
thickness for the respective element or compound can be sent to the
control system. Referring to FIGS. 6 and 7, for a three-stage CIGS
film deposition process, the control system can use previously
established correlation functions of flux rate versus deposition
rate to engage the feedback control loop (200 in FIG. 6, 300 in
FIG. 7) via the flux monitor. At the same time, one or more check
loops can be included to check that the calculated deposition rates
based on the measured flux rates from the flux sensor agree with
previously established source temperatures correlations as well as
associated source power (e.g. current, voltage). For example, the
evaporation rate control system can include a vapor source
temperature sensor and a first check control loop (210 in FIG. 6,
310 in FIG. 7), wherein a correlation function between vapor source
temperature and deposition rate can be used to verify the
deposition rate calculated based on the measured vapor flux rate by
using the vapor source temperature measured by the vapor source
temperature sensor. The evaporation rate control system can also
include a vapor source power sensor and a second check control loop
(220 in FIG. 6, 320 in FIG. 7), wherein a correlation function
between vapor source power and deposition rate can be used to
verify the deposition rate calculated based on the measured vapor
flux rate and/or the vapor source temperature by using the vapor
source power (e.g. current, voltage) measured by the vapor source
power sensor. The entire control loop can be based on self-teaching
algorithms to allow fast selection of the initial target flux rate
as a function of the desired layer thickness.
[0080] Referring to FIG. 6, a multilevel control scheme can use
near infrared reflectometry (NIR) as a means to measure the optical
film thickness of the deposited layer. The multilevel control
scheme in FIG. 7 uses X-ray fluorescence sensor (XRF) as a means to
measure the film thickness of the deposited layer. A film thickness
monitor can use XRF or any other suitable means (e.g. ellipsometry,
transmission, light scattering) to measure the thickness of a
deposited vapor. X-ray fluorescence can measure film thickness and
can further provide compositional information of the deposited
layer, for compounds.
[0081] For faster feedback, the film thickness monitor can be
applied in-situ during the growth phase and can be a second-level
check on film thickness, after vapor flux rate. When the film
thickness monitor detects a discrepancy to the desired target
deposition layer thickness, it can adjust the vapor flux rate and
iterate until the target deposition layer thickness is achieved.
Time delays can be included in the second-level sensors shown in
FIGS. 6 and 7. The timing of the second-level sensor can be after
the vapor flux rate measurement. In an example of a three-stage
CIGS film deposition, XRF feedback can be provided directly
following stage 3 and NIR can be used in-situ in stages 1 and 2.
Stage 3 can allow control of the process in such a way as to
achieve the highest stage 1/stage 3 ratio and Cu-rich excursion
while not requiring in-situ XRF for stage 1 and stage 2. This can
significantly reducing cost and complexity. The timing can be
designed in such a way that the system does not oscillate. In
particular, the response time/time constant of the respective
thermal evaporation source has to be taken into account. Moving
outward from the innermost control loop one has to increase the
time constants for each level of the next outer loop, as otherwise
the system would oscillate.
[0082] Film thickness and substrate temperature can be measured at
any suitable time and any suitable point deposition rate monitoring
process. Film thickness and substrate temperature can be measured
at the same time, or separately, depending on the circumstances. In
some embodiments, the same equipment can be used to measure both
film thickness and substrate temperature and in other embodiments,
different equipment can be used. Non-contacting thermometers or
pyrometers can detect and measure thermal radiation emitted from
the substrate to determine the substrate's temperature in some
embodiments. In other embodiments, thermopiles can be used to
measure the substrate temperature.
[0083] Near infrared reflectometry can be used to measure either
one or both of film thickness and substrate temperature. The near
infrared reflectometer (NIR) can include an active spectral
pyrometry device to extract deposited film thickness information by
measuring and analyzing both the self-emission and reflection of a
surface of the deposited film on the substrate. Where the film
being measured has a thickness greater than the wavelength of the
measuring light, a near infrared reflectometer positioned above the
coated substrate (e.g. on the side with the deposited film) can be
used to measure film thickness and temperature. Where the film
being measured has a thickness less than the wavelength of the
measuring light, a near infrared reflectometer placed above the
coated substrate can be used to measure film thickness and a second
instrument (such as a second near infrared reflectometer) can be
positioned beneath the substrate and directed at the substrate to
obtain temperature data. The near infrared reflectometer can be a
suitable solution for the measurement of moving objects or any
surfaces in harsh conditions that can not be reached or can not be
touched.
[0084] Referring to FIG. 8, near infrared reflectometer 400 can
have an in-situ configuration for in-line deposition process. Near
infrared reflectometer 400 can have lens 410 positioned to receive
thermal radiation 470 from moving substrates 460. Optic fiber
bundle 420 can be used to transmit thermal radiation 460. Mask 430
and filter 440 can be positioned in front of sensor 450. Sensor 450
can be used to measure thermal radiation 460. Sensor 450 can
include an active spectral pyrometry device to extract deposited
film thickness information.
[0085] X-ray fluorescence sensor (XRF) is widely used for elemental
analysis and chemical analysis. X-ray fluorescence sensor (XRF) can
include an energy dispersive spectrometer (EDS). Referring to FIG.
9, the energy dispersive spectrometer can detect the emission of
characteristic "secondary" (or fluorescent) X-rays from a material
that has been excited by bombarding with high-energy X-rays (or
gamma rays). By analyze the emission, the energy dispersive
spectrometer can provide compositional information of the deposited
layer on the substrates.
[0086] In some embodiments, other aspects besides the integral film
thickness may be of importance and additional sensors with
supervisory functions can be introduced into the control
architecture. The embodiments above illustrate the use of sensors
that can help differentiate electronic or optical properties and if
several evaporation sources are used in an in-line configuration
can help establish non-integral compositional information. The
vapor deposition control system can include a control module of
continuous feed thermal evaporation sources. In some embodiments,
the integrated signals from the sensors allow direct access to the
material consumption/mass loss of the source. Therefore, it can be
used to control the feedstock replenishment mechanism controlling
the source to a fixed fill level. In other embodiments, for sources
sensitive to small fluctuations in fill level, the rate monitor and
source power loop can be used to control the continuous feeder.
[0087] A number of embodiments of the invention have been
described. Nevertheless, it will be understood that various
modifications may be made without departing from the spirit and
scope of the invention. It should also be understood that the
appended drawings are not necessarily to scale, presenting a
somewhat simplified representation of various preferred features
illustrative of the basic principles of the invention.
* * * * *